Efficient Structural Parameter Identifiability Analysis for Generator Dynamic Models

This paper proposes an efficient method to perform structural parameter identifiability analysis for generator dy-namic models. Automatic differentiation is applied to efficiently calculate the observability-identifiability matrix. Singular value decomposition is then utilized to determine the rank of the observability-identifiability matrix. In cases of rank deficiency, the results of the singular value decomposition are used for indi-vidual parameter identifiability analysis, further enhancing the computational efficiency. The feasibility of the proposed method is validated on a microbial growth model, and its effectiveness and computational efficiency are further demonstrated on a generator dynamic model with different levels of complexity.